title: "Born Free agentic marketing framework" project: born-free tags: [playbook, framework, ai-agents, marketing-ops, born-free, openclaw] date: 2026-04-30 last_review: 2026-04-30 source: bornfree-hub source_id: docs/marketing/AGENTIC-MARKETING-FRAMEWORK.md upstream_path: /Users/aaronrenecarvajal/bornfree-hub/docs/marketing/AGENTIC-MARKETING-FRAMEWORK.md upstream_created: 2026-02-21 intent: "How OpenClaw + Knowtation agents discover, draft, and follow up with creators while staying inside the Born Free voice gates." depends_on:

• projects/born-free/style-guide/voice-and-boundaries.md
• projects/born-free/outlines/positioning-and-messaging-2026-04.md
• projects/born-free/playbooks/influencer-outreach.md
• projects/store-free/research/public-sources-2026.md

Born Free agentic marketing framework

Vault import of bornfree-hub/docs/marketing/AGENTIC-MARKETING-FRAMEWORK.md (created 2026-02-21). The upstream file is the canonical source. This vault copy lives here so OpenClaw agents (script, social, outreach, blog, etc.) can read it via Knowtation MCP get_note together with the Born Free voice guide, positioning outline, and the outreach playbook. Re-import on upstream changes and re-index. Edits to underlying voice / claims belong in projects/born-free/style-guide/voice-and-boundaries.md, not here.

1. Overview

This framework defines how to use AI agents (Cursor, MCP tools, and external AI) for Born Free marketing — organic influencer outreach, content creation, and ongoing marketing operations.

Core principles

• Human in the loop — you approve outreach, content, and strategy. AI drafts and researches.
• Evidence based — use real data (screenshots, follower counts, engagement) not assumptions.
• Scalable — templates and workflows that work for 10 or 1,000 creators.
• Authentic — messaging reflects "partners, not promoters" — community building, not transactional.

2. Tool stack

Cursor (primary human surface)

• Agent mode — multi-step tasks, code, docs, file edits.
• Chat — quick drafts, Q&A, strategy.
• Rules — bornfree-hub/.cursor/rules/marketing.mdc for consistent marketing context.

MCP tools (when available)

Orchestration (OpenClaw 4.27)

• Codex Computer Use for desktop control (open browser, post to social, upload thumbnail).
• DeepInfra single key for chat, embeddings, image generation, TTS — same DEEPINFRA_API_KEY that hosted Knowtation Hub uses (KNOWTATION_CHAT_PROVIDER=deepinfra), so the LLM bill is unified.
• Cron-driven content agents wake every morning, draft into vault drafts/, then a review dashboard waits for human approve before publish.

External (optional)

• Notion / Airtable — tracker (manual or API).
• Spreadsheet — CSV export for bulk import.

3. Cursor rules for marketing

Maintained in bornfree-hub/.cursor/rules/marketing.mdc. Key rules:

• Positioning: Partners, not promoters — community building, early believers, human adventure.
• No cash offers — Experience Keys, CreditNFTs, Member NFT only.
• Pre-launch — properties coming, faith in platform is the caveat.
• KYC code: BORNFREE100 for free signup.
• Tone: Warm, direct, human. No corporate jargon. Avoid "paid influencer" language — use "partner", "early believer", "support". Emphasize: credits renew forever, DAO governance, community-owned, Strength In Numbers.

Agent-side enforcement: every draft must cite projects/born-free/style-guide/voice-and-boundaries.md path in its output and pass that file's §8 ten-question checklist before promotion to published/.

4. Workflows

A. Influencer discovery from screenshots

Use when sharing screenshots of 0 matches or search results.

• You provide: screenshots of Modash / other platform results.
• Agent analyzes: search terms used, filters applied, suggested alternative queries, free manual discovery paths.
• Agent outputs: refined search strategy + list of free channels to try.

"I'm sharing screenshots of my influencer search. [Attach screenshots]. I got 0 matches. Help me: (1) analyze what might be wrong with my search, (2) suggest alternative search terms and free discovery methods, (3) create a manual discovery checklist I can follow outside paid platforms."

B. Screenshot → contact extraction

• You provide: screenshots of creator profiles (Instagram, YouTube, etc.).
• Agent extracts: name, handle, platform, followers, contact method, content fit notes.
• Agent outputs: structured rows for tracker (CSV or markdown table).

"Extract influencer contact info from these screenshots. For each: name, handle, platform, followers (if visible), email or contact method, content fit. Output as a table I can paste into my tracker."

C. Personalized outreach draft

• You provide: creator name, handle, platform, follower count, content notes.
• Agent drafts: personalized email / DM using templates from projects/born-free/playbooks/influencer-outreach.md.
• You review and send.

"Draft a [email | Instagram DM | Twitter DM] for [creator name] @[handle]. They have [X] followers, focus on [content topic]. Use the partner positioning from projects/born-free/playbooks/influencer-outreach.md and tone from projects/born-free/style-guide/voice-and-boundaries.md. Tier: [1K–10K | 10K–50K | …]."

D. Follow-up sequence

• Agent drafts: day-5 and day-10 follow-ups (shorter, warmer).
• You review and send.

"Draft a [day 5 | day 10] follow-up for [creator]. Original outreach was [email / DM]. Keep it brief, no pressure, one clear CTA."

E. Content creation (general)

• You provide: topic, audience, tone.
• Agent drafts: using Born Free facts from projects/born-free/research/whitepaper.md (when imported) and the public-sources registry.
• You review and publish.

"Draft [content type] for Born Free. Topic: [X]. Audience: [X]. Use facts from projects/store-free/research/public-sources-2026.md and projects/born-free/research/whitepaper.md. Tone: partner / early believer."

5. Task decomposition (agentic)

For complex marketing tasks, break into sub-tasks:

Use the todo_write tool (in Cursor) or OpenClaw's task graph (in production cron jobs) for multi-step tasks. Mark complete as you go.

6. Screenshot analysis workflow

Step 1 — you share

• Screenshots of Modash / other platform (search results, filters, 0 matches).
• Or: creator profiles for contact extraction.

Step 2 — agent analysis

For 0 matches:

• Search terms used.
• Filters that might be too narrow.
• Alternative: "travel creator", "lifestyle influencer", "Web3", "property", "community".
• Free manual discovery paths.

For creator profiles:

• Extract: name, handle, platform, followers, contact.
• Content fit score (1–5).
• Suggested tier.

Step 3 — agent output

• Structured table or CSV.
• Refined search strategy.
• Next actions.

7. Quick reference prompts

8. File structure (cross-repo + vault)

bornfree-hub/docs/marketing/         # source of truth (Git)
├── ORGANIC-INFLUENCER-OUTREACH-PLAN.md
├── AGENTIC-MARKETING-FRAMEWORK.md
└── …

vault/projects/born-free/            # this Knowtation vault (read by agents via MCP)
├── style-guide/voice-and-boundaries.md
├── outlines/positioning-and-messaging-2026-04.md
├── playbooks/
│   ├── influencer-outreach.md       # vault import of upstream plan
│   └── agentic-marketing-framework.md  # this file
└── drafts/                          # OpenClaw agents write here

9. Next steps

1. Verify BORNFREE100 — ensure coupon is active in Stripe / Blockpass.
2. Wire OpenClaw agents to Knowtation hosted MCP (JWT + X-Vault-Id from Settings → Integrations → Hub API). Outreach agent reads projects/born-free/style-guide/voice-and-boundaries.md + this file before drafting.
3. Share screenshots — when ready, attach for agent analysis.
4. Start tracker — Notion, Airtable, or spreadsheet.
5. Pilot first 10 — send 10 outreach messages, refine based on replies, log retro under projects/born-free/research/outreach-retro-<date>.md.